The objective of the project is to test the hypotheses that (1) a mixture of branched chain amino acids, valine, isoleucine, and leucine, or VIL, has a beneficial effect in correcting the interference by excess phenylalanine with central nervous system functioning of patients with phenylketonuria; (2) the improved central nervous system functioning will be reflected in improved performance on standard neuropsychological tests as a consequence of the reduction of brain phenylalanine induced by administration of VIL; and (3) administration of a control mixture consisting of amino acids which have no effect on phenylalanine transport will have no effect on cognitive functioning as measured by neuropsychological test performance. Subjects will be patients with phenylketonuria (PKU) 9 years of age or older in whom adverse behavioral or neurological changes have occurred in association with increased serum phenylalanine concentrations. The subjects may be on phenylalanine restricted diets or unrestricted diets. Prior to the study a neuropsychological test battery will be given for baseline data. VIL or a control amino acid mixture will be given for three month periods. At the end of each three months the neuropsychological test battery will be repeated, and the mixture will be changed. At the end of four study periods of three months each, the data for each individual will be analyzed to compare performance on VIL and on control mixture. The study will be blind to the principal investigator and other professionals involved in the study, the patient and family, and the psychologist. The code for the diet mixtures will be maintained by the nutritionist who prepares and dispenses the mixtures. The study will determine possible practice effects of repeated testing as well as whether VIL promotes improved central nervous system functioning even in the presence of high concentrations of phenylalanine. VIL may represent the first improvement in treatment of PKU since th phenylalanine restricted diet was introduced 30 years ago.